A Review on Hierarchical Clustering-Based Covariance Model to ncRNA Identification

Recent discoveries have revealed that functional discovery of noncoding RNAs (ncRNAs) has gradually acquired attention among researchers in bioinformatics domain. ncRNA families are believed to be responsible for a variety of biological functionalities, ranging from gene expression regulation to catalytic activities, when the others are still to be unveiled. These new recoveries have opened many aspects in ncRNA research, for example in functional subgroups discovery. Hence, cross fertilization solutions originated from computational intelligence concepts and algorithms has started to achieve promising results. For instance, data clustering is one of the popular techniques in many different domains for the purpose of Covariance Model (CM) in ncRNA identification. Hierarchical clustering is the most frequently used mathematical technique to group a set of ncRNAs in human into different families based on sequence similarity. However, conventional algorithms have some shortcomings such as the sequence structures of each family will be significantly diluted when the number of sequence features for known family dataset increases. This study presents a literature review on the hierarchical clustering algorithm and its variants for ncRNA family identification using the sequence structure.